TensorFlow学习之 mnist_deep.py

源代码

mnist_deep.py的网络结构定义如下

def deepnn(x):
  """deepnn builds the graph for a deep net for classifying digits.

  Args:
    x: an input tensor with the dimensions (N_examples, 784), where 784 is the
    number of pixels in a standard MNIST image.

  Returns:
    A tuple (y, keep_prob). y is a tensor of shape (N_examples, 10), with values
    equal to the logits of classifying the digit into one of 10 classes (the
    digits 0-9). keep_prob is a scalar placeholder for the probability of
    dropout.
  """
  # Reshape to use within a convolutional neural net.
  # Last dimension is for "features" - there is only one here, since images are
  # grayscale -- it would be 3 for an RGB image, 4 for RGBA, etc.
  with tf.name_scope('reshape'):
    x_image = tf.reshape(x, [-1, 28, 28, 1])

  # First convolutional layer - maps one grayscale image to 32 feature maps.
  with tf.name_scope('conv1'):
    W_conv1 = weight_variable([5, 5, 1, 32])
    b_conv1 = bias_variable([32])
    h_conv1 = tf.nn.relu(conv2d(x_image, W_conv1) + b_conv1)

  # Pooling layer - downsamples by 2X.
  with tf.name_scope('pool1'):
    h_pool1 = max_pool_2x2(h_conv1)

  # Second convolutional layer -- maps 32 feature maps to 64.
  with tf.name_scope('conv2'):
    W_conv2 = weight_variable([5, 5, 32, 64])
    b_conv2 = bias_variable([64])
    h_conv2 = tf.nn.relu(conv2d(h_pool1, W_conv2) + b_conv2)

  # Second pooling layer.
  with tf.name_scope('pool2'):
    h_pool2 = max_pool_2x2(h_conv2)

  # Fully connected layer 1 -- after 2 round of downsampling, our 28x28 image
  # is down to 7x7x64 feature maps -- maps this to 1024 features.
  with tf.name_scope('fc1'):
    W_fc1 = weight_variable([7 * 7 * 64, 1024])
    b_fc1 = bias_variable([1024])

    h_pool2_flat = tf.reshape(h_pool2, [-1, 7*7*64])
    h_fc1 = tf.nn.relu(tf.matmul(h_pool2_flat, W_fc1) + b_fc1)

  # Dropout - controls the complexity of the model, prevents co-adaptation of
  # features.
  with tf.name_scope('dropout'):
    keep_prob = tf.placeholder(tf.float32)
    h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob)

  # Map the 1024 features to 10 classes, one for each digit
  with tf.name_scope('fc2'):
    W_fc2 = weight_variable([1024, 10])
    b_fc2 = bias_variable([10])

    y_conv = tf.matmul(h_fc1_drop, W_fc2) + b_fc2
  return y_conv, keep_prob

其中几乎每一层，都包含

  with tf.name_scope('reshape'):
    x_image = tf.reshape(x, [-1, 28, 28, 1])

语句。下面来仔细探究下tf.name_scope('reshape')的作用。

解析之 tf.name_scope('reshape')

在TensorFlow中，每一个定义的变量实际上是有名字(name)的。

例如: 代码

import tensorflow as tf
var1 = tf.Variable(tf.zeros([784,10]))
print(var1.name) #输出为 Variable:0

这段代码的输出为'Variable:0'即为变量var1的名字(var1.name)，这个名字是系统默认分配的。

也可以在创建变量时，自己给定变量名，即：

import tensorflow as tf
var1 = tf.Variable(name = 'var1', initial_value = tf.zeros([784,10]))
print(var1.name) #输出为 var1:0

这时代码的输出为自己指定的名字'var1:0'

而使用tf.name_scope('reshape')命令，可以指定一段代码中所有变量的名称前缀。

例如

with tf.name_scope("reshape"):
    var1 = tf.Variable(name = 'var1', initial_value = tf.zeros([784,10]))
    print(var1.name) #输出为 reshape/var1:0 

这时，输出变成了'reshape/var1:0' 。也就是说，tf.name_scope('reshape') 命令，在var1的名字前面加上了当前命名空间的名字 'reshape' 。

所以说代码中加入的with tf.name_scope('reshape'): 是为了命名的直观考虑的

也即是说，代码

x = tf.placeholder(tf.float32, [None, 784])
with tf.name_scope('reshape'):
    x_image = tf.reshape(x, [-1, 28, 28, 1])

  # First convolutional layer - maps one grayscale image to 32 feature maps.
with tf.name_scope('conv1'):
    W_conv1 = weight_variable([5, 5, 1, 32])
    b_conv1 = bias_variable([32])
    h_conv1 = tf.nn.relu(conv2d(x_image, W_conv1) + b_conv1)

  # Pooling layer - downsamples by 2X.
with tf.name_scope('pool1'):
    h_pool1 = max_pool_2x2(h_conv1)

  # Second convolutional layer -- maps 32 feature maps to 64.
with tf.name_scope('conv2'):
    W_conv2 = weight_variable([5, 5, 32, 64])
    b_conv2 = bias_variable([64])
    h_conv2 = tf.nn.relu(conv2d(h_pool1, W_conv2) + b_conv2)

  # Second pooling layer.
with tf.name_scope('pool2'):
    h_pool2 = max_pool_2x2(h_conv2)
    
print(x_image.name)
print(W_conv1.name)
print(b_conv1.name)
print(h_pool2.name)
print(W_conv2.name)
print(b_conv2.name)
print(h_pool2.name)

的输出是

reshape/Reshape:0
conv1/Variable:0
conv1/Variable_1:0
conv1/Relu:0
pool2/MaxPool:0
conv2/Variable:0
conv2/Variable_1:0
conv2/Relu:0
pool2/MaxPool:0

而如果不加上with tf.name_scope('reshape'): ，也即代码

import tensorflow as tf
from mnist_deep import *
x = tf.placeholder(tf.float32, [None, 784])

x_image = tf.reshape(x, [-1, 28, 28, 1])

# First convolutional layer - maps one grayscale image to 32 feature maps.
W_conv1 = weight_variable([5, 5, 1, 32])
b_conv1 = bias_variable([32])
h_conv1 = tf.nn.relu(conv2d(x_image, W_conv1) + b_conv1)

  # Pooling layer - downsamples by 2X.
h_pool1 = max_pool_2x2(h_conv1)

  # Second convolutional layer -- maps 32 feature maps to 64.
W_conv2 = weight_variable([5, 5, 32, 64])
b_conv2 = bias_variable([64])
h_conv2 = tf.nn.relu(conv2d(h_pool1, W_conv2) + b_conv2)

  # Second pooling layer.
h_pool2 = max_pool_2x2(h_conv2)
    
print(x_image.name)
print(W_conv1.name)
print(b_conv1.name)
print(h_conv1.name)
print(h_pool2.name)
print(W_conv2.name)
print(b_conv2.name)
print(h_conv2.name)
print(h_pool2.name)

的输出将是

Reshape:0
Variable:0
Variable_1:0
Relu:0
MaxPool_1:0
Variable_2:0
Variable_3:0
Relu_1:0
MaxPool_1:0

一团糟是吧，那个变量是干啥的不好分清楚。